The kidney is a particularly rich source for prostaglandins and prostaglandins modulate a broad range of renal functions including microvascular hemodynamics, renin release and tubular salt and water reabsorption. The normal adult mammalian kidney expresses COX2 and cyclooxygenase metabolites and they have been implicated in functional and structural alterations in glomerular and tubulointerstitial inflammatory diseases. In experimental animal models of kidney disease, COX2 selective inhibitors decreased proteinuria and inhibited the development of glomerulosclerosis. In addition, the use of COX2 inhibitors decreased the expression of TGF-beta1 in Type III and IV collagen. Clinical studies in humans have minimally examined the renal effect of the COX2 inhibitors. They have been demonstrated to decrease the urinary excretion of prostaglandins, reduce urinary sodium excretion and have conflicting results on the acute changes in GFR. Nonsteroidal anti-inflammatory agents which work less specifically than COX2 inhibitors in the kidney have been demonstrated to decrease proteinuria in subjects with refractory nephrotic syndrome. In this study, we propose to extend the animal and human data suggesting a potential beneficial effect of COX2 inhibitors in renal disease as manifested by its acute hemodynamic effects decreasing proteinuria and perhaps initially single nephron GFR. In an ongoing pilot study, subjects with diabetic nephropathy manifested by proteinuria and declining renal function on inhibitors of the renin angiotensin system have been randomly assigned in a crossover design to COX2 inhibitors, washout, placebo or the reverse. The COX2 inhibitor being used for this pilot study is Celebrex 200 mg/day. Recognizing there may be safety concerns in regards to hyperkalemia and acute renal failure the safety monitoring committee has followed the 12 subjects randomized thus far and found the safety profile excellent. We now propose to extend this pilot study to complete the 30 proposed subjects. We propose to use a decrease in proteinuria as a surrogate for renal function in this short term pilot study. A decline in proteinuria in this pilot study would support consideration of a long term trial to examine the efficiency of COX2 inhibitors in preserving renal function. Also the pilot study allows us to establish procedures, better estimate recruitment goals and establish safety for a larger trial.